Date: Mon, 19 Sep 2005 14:40:56 +0300 From: AK <lesha@ns.divo.ru> To: Chuck Swiger <cswiger@mac.com> Cc: freebsd-questions@freebsd.org Subject: Re: "panic: getnewbuf: locked buf" on heavy load 5.4-STABLE Message-ID: <200509191440.56774.lesha@ns.divo.ru> In-Reply-To: <432AF207.6030908@mac.com> References: <200509161254.14108.lesha@ns.divo.ru> <200509161909.10463.lesha@ns.divo.ru> <432AF207.6030908@mac.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 16 September 2005 19:25, Chuck Swiger wrote: > > To get a crashdump I should have swapspace >= total ram. > > Currently it is only 2G, I'm trying to think of smth. to get more > > swapspace. dmesg is available @ http://getfile.biz/dmesg.txt > > Take a look at /boot/loader.conf, you can limit the amount of physical RAM > FreeBSD uses to less than 2GB for long enough to get a valid dump, assuming > the crash is easily reproducable. Well... it is easy reproduceable, but .... with debugging kernel system is not going into panic, it just freezes. When I limit apache to 400 connections (MaxClients), it can work for ages, setting it to 500-600 will crash system in a few hours, setting it to 1000 will crash system as soon as people starting to download something. (when enough users connect) I have tried starting apache with MaxClients 1500, connecting my laptop with crosscable, and starting benchmarks/siege on the laptop with 750 simultaneous connections. Server dies in less than 1 secons. It just freezes. It can't be network card, as I have tried different ones. It is not RAM, I have changed it already. What else can it be? Alex.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200509191440.56774.lesha>